THREE ESSAYS ON ANTIPOVERTY POLICIES IN THE UNITED STATES

This dissertation is comprised of three empirical studies considering two major antipoverty policies in the United States, the Earned Income Tax Credit (EITC) and the Supplemental Nutrition Assistance Program (SNAP, formerly known as the Food Stamp Program). The first chapter finds that a major federal expansion of the EITC reduced material hardship for families with children across multiple dimensions. The second tests several hypothesized predictors of EITC adoption by states and finds that public attitudes about the deservingness of Black Americans significantly predicts state EITC adoption. The final chapter aims to quantify the contribution of state policy choices to racial disparities in SNAP participation and food insecurity. I find that SNAP-eligible Hispanic households are disproportionately exposed to administrative burden in SNAP, but overall results suggest that factors outside of administrative burden are the driving force behind racial gaps in program participation and food security. Taken together, these chapters tell a story of how state and federal policymaking shape the experience of having low income in the United States. They also consider how the social infrastructure in the United States distributes benefits and barriers unevenly, often along lines of race and class (Parolin, 2021; Gaines et al., 2021; Herd et al., 2023). In my first chapter, I pool data from four waves of the Survey of Income and Program Participation (SIPP) and use a difference-in-difference identification strategy to estimate the effect of a 2009 expansion to the federal EITC for families with three or more children on the likelihood of experiencing material hardship, the number of hardships reported, and six individual dimensions of material hardship: hardship related to essential expenses, housing, utilities, and health care, food insecurity, and food insufficiency. While the effects of the EITC on poverty and employment are well-established, less is known about the effect of the EITC on experiences of material hardship. Material hardship is a distinct measure of household-level economic wellbeing from income and poverty status and represents challenges to meeting basic needs like food, housing, and health care (Heflin, 2014). Prior work has shown that material hardship decreases monotonically with income (Sullivan, Turner, and Danziger, 2008), though correlations between measures of hardship and measures of income poverty are moderate to weak (Mayer and Jencks, 1989; Iceland and Bauman, 2007; Sullivan, Turner, and Danziger, 2008). My findings indicate that the 2009 EITC expansion for families with three or more children led to a decrease in the likelihood of experiencing five of the six types of material hardship, the likelihood of experiencing greater than zero hardships, and the total number of hardships. If, in addition to lifting families out of poverty, a more generous EITC reduces experiences of material hardship, and there is evidence that state EITCs mirror the antipoverty and public health benefits of the federal EITC, why had fewer than half of American states adopted a state EITC as of 2016? Is the pattern of non-adoption simply a function of available resources, or ideological preferences for redistribution? The study of policy innovation and diffusion helps researchers understand the factors that influence adoption and diffusion of a policy (Mooney, 2020). My second chapter expands the policy diffusion literature by drawing upon theories of “deservingness” and studying the previously unexamined policy area of state EITCs. In addition to traditionally theorized and empirically tested predictors of policy adoption – available resources, ideology, a shared border or region with past adopters, or simply time passed since the first adoption – this paper presents the first use of a time-varying state-level measure of racial resentment in a policy diffusion study. Following Lanford and Quadagno (2022), I argue that Kinder and Sanders’ (1996) Racial Resentment Scale (RRS) serves as a plausible measure of perceived deservingness of Black Americans. Comparing the predictive value of this variable to that of racial threat as measured by demographic data, provides support for a new, more sophisticated option for assessing the influence of racialized public opinion on adoption of policies that are shown to enhance economic equity. I find that low anti-Black affect significantly predicts adoption of state EITCs and my findings highlight the Racial Resentment Scale (RRS) as a more sophisticated measure of racialized attitudes compared to the commonly used construct of racial threat. Finally, I consider racial differences in program participation and food security within the SNAP-eligible population. There is evidence that SNAP reduces racial disparities in food insecurity for participants and that administrative burdens predict SNAP participation (Fox et al., 2023; Samuel et al., 2023; Brantley et al., 2020). However, there have been no studies designed to detect the contribution of burdensome policies to racial disparities in SNAP participation. Administrative burden, or the learning, compliance, and psychological costs imposed on citizens to access public goods or services, modifies the ease of applying for and maintaining access to benefits and has a distributive effect on antipoverty policies, including SNAP (Herd and Moynihan, 2018). When inequitable experiences of burden differ by race, they are called racialized burdens (Ray et al., 2023). Racialized burdens occur when facially race-neutral policies are structured such that historically disadvantaged groups disproportionately face uncertainty, stigma, or risk of breaking a rule, while trying to receive benefits for which they are eligible. The last chapter of my dissertation uses a decomposition framework to examine racial differences in food insecurity conditional on SNAP eligibility, and specifically consider to what degree the presence of burdensome policies explains these differences. The findings suggest that reducing administrative burden in SNAP may increase program participation for Hispanic households and reduce transaction costs faced by all households but may not significantly narrow racial gaps in food security.